Background and Purpose Cognitive reserve is important for the individual susceptibility to dementia. Among the various determinants of cognitive reserve, the number of years of formal education would be of prime importance. We performed this study to explore its contribution to the clinical characteristics of dementia. Methods We included consecutive mild Alzheimer's disease (AD) dementia patients (clinical dementia rating, CDR=0.5 or 1) who visited our memory clinic and were older than 70 years at the evaluation from October 2013 to September 2015. According to the number of years of education, the corresponding subjects was grouped into two extreme educational groups, low education (illiterate or partially illiterate, LE, n=43) vs. high education (more than 9 years of education, HE, n=34). Among these patients, we compared various demographic, neuropsychological and neuroimaging characteristics. Results The groups were comparable in terms of age, frequency of hypertension and diabetes, and CDR and its sum of box. However, female gender was more frequent in the LE group (p=0.000). Also this group showed a more depressive mood on the geriatric depression scale (p=0.007). The raw scores on Korean Version of Mini-Mental State Examination and well-validated neuropsychological tests were lower in the LE group. However, the cognitive performance was found to be more preserved in the LE group than in the HE group when assessed using the z-score in certain specified tests on univariate and multivariate analyses (p<0.05). The Schelten's grade of medial temporal atrophy was similar between the two educational groups. Also, the degree of combined ischemic burden did not differ between the two groups. Conclusions We identified that the prevalence of depressive mood and the extent of decline from the corresponding norm in a particular neuropsychological performance differ according to the educational level of mild AD patients.